Landmark Melanoma Trial Does Not Erase Controversies

Nick Mulcahy

February 12, 2014

The common practice of sentinel-lymph node (SLN) biopsy in patients with melanoma does not improve their long-term survival, according to the final results of a landmark international trial published today in the New England Journal of Medicine.

The investigators of the Multicenter Selective Lymphadenectomy (MSLT-I) trial compared SLN biopsy with a watch-and-wait approach (only removing nodes once palpable) in melanoma.

The 10-year, disease-specific survival in the overall study population was the primary outcome and was not significantly different between the biopsy and observation arms (81.4% vs 78.3%; P = .18)

However, the investigators said that the lack of a difference was "unsurprising" because event rates were lower than expected. On the other hand, SLN biopsy can prolong disease-free survival, they report.

"If lymph node metastases are present, it is much, much better to have them removed early, while they are still microscopic," said senior author Mark Faries, MD, in an email to Medscape Medical News. He is from the John Wayne Cancer Institute, in Santa Monica, California.

Overall, the study findings provide enough evidence for clinicians to use SLN biopsy routinely in many patients, the authors say.

"These long-term results clearly validate the use of sentinel node biopsy in patients with intermediate-thickness and thick primary melanomas," write the authors, led by the recently deceased sentinel-node pioneer Donald Morton, MD, also from the John Wayne Cancer Institute.

A pair of editorialists also suggested that the lack of a disease-specific survival benefit was not an impediment to endorsing the use of SLN biopsy.

"The MSLT-I results strongly support the continued use of sentinel biopsy in surgical practice," write Charles Balch, MD, of the University of Texas Southwestern Medical Center, in Dallas, and Jeffrey E. Gershenwald, MD, of the University of Texas MD Anderson Cancer Center, in Houston.

But one study critic strongly disagrees with these conclusions and endorsements.

The use of SLN biopsy in the trial "offers no survival advantage of any variety," said J. Meirion Thomas, MS, FRCP, FRCS, chair of surgical oncology at Imperial College, London, United Kingdom.

"I repeat the charge that sentinel node biopsy in melanoma benefits doctors and hospitals but not patients, who may be considerably harmed by this intervention," he told Medscape Medical News in an email. "It is a staging procedure and nothing more."

It is a staging procedure and nothing more.

Dr. Thomas is a long-time critic of the routine use of SLN biopsy and of the MSLT-I trial.

But he is not alone in interpreting the results of the trial much differently than the authors and editorialists.

"There is a little signal from this trial that you will do better if you undergo sentinel node biopsy and have intermediate-thickness melanoma," Daniel Coit, MD, told Medscape Medical News in an interview. He is from Memorial Sloan Kettering Cancer Center, in New York City, and is a chair of the National Comprehensive Cancer Network (NCCN) melanoma panel.

Unfortunately, it is impossible to know which small fraction of intermediate-thickness patients might benefit from the procedure when "they first walk through your door," he added.

Nonetheless, Dr. Coit routinely uses SLN biopsy with his patients and believes the new results reinforce its importance in staging and prognosis. "It is clearly the most important prognostic factor," he said, referring to the absence or presence of cancer in the lymph nodes in the study patients. "But it does not have an impact on survival."

Dr. Coit also acknowledged that there are 2 outstanding downsides to SLN biopsy in this setting: complications and cost. While the new paper does not include either factor in its scope, he believes it is important to say that, while SLN biopsy is priced fairly given the resources employed, it "costs a lot for a staging test."

Dr. Coit also emphasized his gratitude to the late and "revered" Dr. Morton, who, along with colleagues, is credited with having introduced sentinel-lymph node biopsy in 1992 as an alternative to simply removing all of a melanoma patient's nodes.

There are legitimate criticisms of the trial but no concerns about the man.

"There are legitimate criticisms of the trial but no concerns about the man," Dr. Coit said. "He not only generated an idea but tested it rigorously."

Survival Benefit in Subset of Patients Reviewed

The investigators randomized 2001 patients, all of whom underwent wide local excision, to receive either sentinel-node biopsy (SLN biopsy group, 60% of patients) or postoperative nodal observation (observation group, 40% of patients).

Observation-group patients received delayed lymphadenectomy if metastases were detected in the regional lymph node basin.

The authors reported their 10-year results for the 1347 patients with intermediate-thickness (1.2 mm–3.5 mm) and 314 with thick (> 3.5 mm) primary melanomas. Patients with thin (< 1.2 mm) melanomas did not have enough events to be included in the analysis.

The authors summarize their most important findings in their final sentence of their new paper.

"The procedure provides accurate and important staging information…" they start off.

There is no dispute here. The trial confirms the value of SLN biopsy as an indicator of prognosis, agreed both Dr. Thomas and Dr. Coit.

The results indicate that sentinel-node status was the most important prognostic factor for 10-year melanoma-specific survival of patients with intermediate or thick melanoma. And multivariate analysis confirmed it.

Take the example of the intermediate-thickness patients. The 10-year survival rate was 62.1% for those with positive nodes and 85.1% for those with tumor-free nodes (P < .001).

The authors continue their all-important final sentence by saying that (the procedure) "enhances regional disease control…"

This refers to the findings that the 10-year disease-free survival rates were significantly improved in the biopsy group, as compared with the observation group, among patients with intermediate-thickness melanomas (71.3% vs 64.7%; P = .01), and those with thick melanomas (50.7% vs 40.5%; P = .03).

Dr. Coit believes that this finding is not as strong as it might appear. "A skeptic could argue that 'I can get regional control later [when a node becomes palpable]." The implication is that the later removal of lymph nodes will not necessarily cost a patient any survival time.

Dr. Thomas also commented that the finding is not impressive because the "disease-free survival advantage in the biopsy arm was an inevitable consequence of trial design and did not reflect any therapeutic advantage for patients with melanoma."

The authors then continue — and conclude — their final sentence by saying that (the procedure) "among patients with nodal metastases, appears to improve melanoma-specific survival substantially."

Why did the authors use the word "appears" and not some stronger, more definitive term? "We were just being cautious [perhaps overly cautious] in observing this benefit," Dr. Faries said in his email.

Here, the authors discuss the finding that, among patients with intermediate-thickness melanomas, the 10-year melanoma-specific survival rate was 62.1% in the biopsy group vs 41.5% in the observation group (P = .006) — a dramatic 20% absolute improvement.

This treatment-related difference was not, however, seen for patients with thick melanomas.

Thus, there was a survival benefit seen with SLN biopsy in the study after all, the authors assert, but only in the subset of patients who had intermediate-thickness melanomas and who had positive lymph nodes.

Dr. Coit explained what that meant in simple and practical terms.

If you saw 100 intermediate thickness patients, you could expect 20% (n = 20) to have a positive SLN biopsy. Out of that group, you could expect 20% (n = 4) to have a better outcome (improved melanoma-specific survival). Thus for every 100 patients (but only the ones with intermediate-thickness melanoma) who undergo SLN biopsy, you might have 4 with improved disease-specific survival. But there is no way of knowing who exactly will benefit and this survival analysis was not the primary outcome of the trial, Dr. Coit commented further.

Dr. Thomas was stronger in his criticism here, saying this last analysis is "statistically invalid" because a postrandomization, selected, subgroup analysis is only valid when the patients concerned can be identified and stratified before randomization. And that is not what happened in this case.

Furthermore, the authors went ahead and published this finding even though Dr. Thomas successfully appealed to the National Institutes of Health to not have it included in the study, he said.

"The failure to accurately report the results of MSLT-I brings shame on the authors and indeed on the Journal," Dr. Thomas said in his email.

Dr. Faries said that more will be learned about the utility of SLN biopsy in the future, as another major trial (MSLT-II) accrues patients. "Most patients with SLN involvement don't have other involved nodes," he explained. This begs the question of whether or not additional completion procedure (which currently follows a positive SLN in standard practice) is necessary. The MSLT-II trial randomizes patients with SLN metastases to completion dissection or nodal observation with exams and ultrasound.

Supported by a grant from the National Cancer Institute with additional support at Australian centers provided by the Australian and New Zealand Melanoma Trials Group.

N Engl J Med . 2014;370:599-609 and 663-664.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.